Is there any way to use glossary module in combination with free tagging. Today, If i create a special "glossary" node-type and making a new term using free tagging, i still have to open that term and fill in a description to get that word "glossarized".

I mean, would it not be good if the glossary module gets the description from somewhere in the "glossary" node-type. Maybe from the first x characters of text that is shown in some views (dont know a better way to describe what i mean).

Comments

nancydru’s picture

Status: Active » Closed (works as designed)

There is no "glossary" node type. Do you mean from the node on which the term was first used? Or the node on which it was most recent used?

In actuality, unless they are being used for a "Detailed description" nodes should not be tagged with the glossary vocabulary terms.

jvlagsma’s picture

I've been having a play with this. I want to use a free tagging vocabulary for the glossary module and attach nodes to its categories. When clicking on the glossary term I want a list of nodes with that term. The part where this does not work is where glossary module says "// Omit terms without a description. those are usually container terms." as my free tagging categories never get descriptions (unless I manually add them all, not practical).

I've disabled the if statement per below and it now works like I want it to:

//        if ($term->description) {
          $term->synonyms = $synonyms[$term->tid];
          $term->synonyms[] = $term->name;
          $term->vid = $vid;
          $terms[] = $term;
//        }

Could this maybe become a setting?

nancydru’s picture

Assigned: Unassigned » nancydru
Status: Closed (works as designed) » Fixed

Okay, you have a setting.

Fix committed.

jvlagsma’s picture

Wonderful, many thanks!
I've been using glossary for as long as I run my drupal site (3yrs) but this fix gives it a totally different perspective, it makes the module dynamic and tag driven, it puts my community in the driver seat. I'm using Unitag to get moderated new tags from users and Im using taxonomy manager to merge, manage, delete tags as appropriate, works great.
However, with my set of tags having increased from 70 to 300 I'm noticing a performance impact, pagetimes have increased from 2.5 sec to well over 8 sec. When switching off glossary I'm down to 2.2 sec so I've left it switched off 'for the time being and am now looking for a way to solve the problem.
The glossary docs say the module is using cache_filter but I'm not seeing this table being filled, it always has 0 rows. Also the settings page says my input format is not eledgable for caching eventhough this cache_filter is an always on drupal standard cache (I'm on drupal 5.6). So I'll be doing some more testing, if anyone can share a pointer that would be greatly appreciated!
Thanks for looking after Glossary

nancydru’s picture

Actually it's the core filter module that uses the cache_filter table. Any filter module has the capability to force the cache off. I've seen my Full HTML format get turned off and I don't know why. I can turn it back on and it will stay on for a while then it gets turned back off. I do not have this problem with the Filtered HTML format.

At the top of each filter format settings page it tells you whether that format is being cached. If it isn't, you can go into the filter_format table and try turning it on. It would help if you could also tell me which other filter modules are being used (I suspect the PHP filter).

For purposes of issue tracking I would appreciate you opening this as a new issue (probably support request to start with). The way Drupal is set up, tracking more than one issue in a thread is very difficult. Thanks.

nancydru’s picture

I'd still like to see the performance problem opened as a new issue. If you have the Site Documentation module, it would help to include the output of the "Input Formats and Filters" section.

jvlagsma’s picture

Been busy with one of the moderators deleting the entire forum from the site.. ah well, the good news is it gave me two attempts at the performance issue. First, by disabeling one of the filters (html corrector) the cache_filter started working. Switching html corrector on again still kept the cache working. Second attempt html corrector didn't make a difference, but this time highlight did the trick. Anyway, I can confirm that with the cache running performance is acceptable. Good on repeated reads but with sometimes longer delays when glossary needs to do its work. I agree it feels like there is inconsistency in how this cache behaves and this needs a separate issue, most probably against core.. but how to phrase the problem.. I cannot really reproduce the circumstances. I'll have a stab at my test system first, hopefully I can recognise some sort of a pattern.
Thanks
Johan

nancydru’s picture

On your test system, you can go into the filter_formats table and turn off the cache setting. That should give you more repeatable results and better measurements. One benchmark that I would be interested in is case-sensitive versus case-insensitive.

But for your request against core, I suspect you're looking for something along the lines of vocabulary and term caching. Unfortunately, it's going to take an enormous amount of convincing and benchmarking to get them to listen to you. Vocabularies are "sort of" cached already, but I don't think terms are. Such practice could easily become quite a memory hog on some sites with huge vocabularies (like locate lists).

BTW, I did make a small tweak to the Site Doc module for this (showing filters that set "no cache"). It should be available after midnight GMT.

Anonymous’s picture

Status: Fixed » Closed (fixed)

Automatically closed -- issue fixed for two weeks with no activity.